SX-Chaos: Full Lyapunov Spectrum Computation via SIMD Vectorization on Modern Hardware
Fuente:
Zenodo
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Lenguaje: | inglés |
| Publicado: |
Zenodo
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901469209296896 |
|---|---|
| author | Pirolo, Andrés Sebastián |
| author_facet | Pirolo, Andrés Sebastián |
| contents | <blockquote> <p dir="ltr">The future of complex dynamic modeling and real-time computation lies at the edge, yet high-dimensional continuous systems have traditionally required costly, power-hungry HPC (High-Performance Computing) or datacenter infrastructure.</p> <p dir="ltr"><strong>Now</strong>, the <strong>SX-Chaos</strong> engine challenges this structural limitation. This repository demonstrates a fundamental algorithmic breakthrough: collapsing the traditional \mathcal{O}(N^2) computational bottleneck down to \mathcal{O}(N). By coupling this architectural restructuring with aggressive, edge-native SIMD vectorization, we unlock dormant processing power in standard consumer hardware.</p> <p dir="ltr"><strong>The Result:</strong> Sub-second execution of 40-dimensional chaotic spectrums (processing 50,000 integration steps) on a single consumer-grade mobile core. This achieves a performance-per-watt ratio that fundamentally disrupts traditional scaling models, making the marginal cost of this compute power <strong>almost free</strong> compared to renting datacenter instances.</p> <p dir="ltr"><strong>Market Implications:</strong> This establishes the viability of running complex, high-dimensional simulations directly on edge devices (smartphones, IoT arrays, autonomous systems, sensor networks) entirely offline. By eliminating recurring cloud compute costs and communication latency, real-time advanced modeling is <strong>now</strong> available anywhere.</p> <p dir="ltr"><strong>Explore the potential:</strong></p> <ul> <li>Read the PiroloSX_Chaos_Pirolo_2026.pdf manuscript for comprehensive benchmarks, methodology, and empirical proof of the performance leap.</li> <li>Review the PiroloSXChaos_L96_Release.zip package for technical validation.</li> </ul> <p dir="ltr"><em>Note: This public release is strictly governed by the PolyForm Noncommercial License 1.0.0. Industrial application, hardware synthesis (FPGA/ASIC), enterprise AI training, or commercial deployment requires a separate licensing agreement with the author.</em></p> </blockquote> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18868069 |
| institution | Zenodo |
| language | eng |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | SX-Chaos: Full Lyapunov Spectrum Computation via SIMD Vectorization on Modern Hardware Pirolo, Andrés Sebastián Lyapunov exponents Kaplan-Yorke dimension Chaotic dynamical systems SIMD vectorization Lorenz-96 <blockquote> <p dir="ltr">The future of complex dynamic modeling and real-time computation lies at the edge, yet high-dimensional continuous systems have traditionally required costly, power-hungry HPC (High-Performance Computing) or datacenter infrastructure.</p> <p dir="ltr"><strong>Now</strong>, the <strong>SX-Chaos</strong> engine challenges this structural limitation. This repository demonstrates a fundamental algorithmic breakthrough: collapsing the traditional \mathcal{O}(N^2) computational bottleneck down to \mathcal{O}(N). By coupling this architectural restructuring with aggressive, edge-native SIMD vectorization, we unlock dormant processing power in standard consumer hardware.</p> <p dir="ltr"><strong>The Result:</strong> Sub-second execution of 40-dimensional chaotic spectrums (processing 50,000 integration steps) on a single consumer-grade mobile core. This achieves a performance-per-watt ratio that fundamentally disrupts traditional scaling models, making the marginal cost of this compute power <strong>almost free</strong> compared to renting datacenter instances.</p> <p dir="ltr"><strong>Market Implications:</strong> This establishes the viability of running complex, high-dimensional simulations directly on edge devices (smartphones, IoT arrays, autonomous systems, sensor networks) entirely offline. By eliminating recurring cloud compute costs and communication latency, real-time advanced modeling is <strong>now</strong> available anywhere.</p> <p dir="ltr"><strong>Explore the potential:</strong></p> <ul> <li>Read the PiroloSX_Chaos_Pirolo_2026.pdf manuscript for comprehensive benchmarks, methodology, and empirical proof of the performance leap.</li> <li>Review the PiroloSXChaos_L96_Release.zip package for technical validation.</li> </ul> <p dir="ltr"><em>Note: This public release is strictly governed by the PolyForm Noncommercial License 1.0.0. Industrial application, hardware synthesis (FPGA/ASIC), enterprise AI training, or commercial deployment requires a separate licensing agreement with the author.</em></p> </blockquote> |
| title | SX-Chaos: Full Lyapunov Spectrum Computation via SIMD Vectorization on Modern Hardware |
| topic | Lyapunov exponents Kaplan-Yorke dimension Chaotic dynamical systems SIMD vectorization Lorenz-96 |
| url | https://doi.org/10.5281/zenodo.18868069 |